data/analysis/clothing_precision_trial_compare_20260322T113759Z.json

{
  "timestamp_utc": "2026-03-22T11:39:34.040448Z",
  "subset_size": 10,
  "subset_ids": [
    37076,
    203825,
    335343,
    1023509,
    2141904,
    2291202,
    3285630,
    260449,
    1325009,
    212271
  ],
  "baseline_file": "E:\\image\\backup\\Prompt_Squirrel_RAG\\data\\eval_results\\model_ab_n30_mistral_small2501_20260322.jsonl",
  "new_file": "E:\\image\\backup\\Prompt_Squirrel_RAG\\data\\eval_results\\model_ab_n10_mistral_small2501_clothing_precision_20260322T113759Z.jsonl",
  "usage_before_usd": 4.796378237,
  "usage_after_usd": 4.797453587,
  "usage_delta_usd": 0.0010753499999998084,
  "baseline_clothed": {
    "tp": 4,
    "fp": 6,
    "fn": 0,
    "precision": 0.4,
    "recall": 1.0,
    "f1": 0.5714285714285715
  },
  "baseline_clothing": {
    "tp": 4,
    "fp": 6,
    "fn": 0,
    "precision": 0.4,
    "recall": 1.0,
    "f1": 0.5714285714285715
  },
  "new_clothed": {
    "tp": 4,
    "fp": 6,
    "fn": 0,
    "precision": 0.4,
    "recall": 1.0,
    "f1": 0.5714285714285715
  },
  "new_clothing": {
    "tp": 4,
    "fp": 6,
    "fn": 0,
    "precision": 0.4,
    "recall": 1.0,
    "f1": 0.5714285714285715
  },
  "avg_f1_baseline": 0.5784,
  "avg_f1_precision_prompt": 0.58621,
  "per_id": [
    {
      "id": 37076,
      "baseline": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.7222,
        "R": 0.8667,
        "F1": 0.7879
      },
      "precision_prompt": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.7778,
        "R": 0.9333,
        "F1": 0.8485
      },
      "added_tags": [
        "bell",
        "blue_eyes"
      ],
      "removed_tags": [
        "better_version_at_source",
        "collar"
      ]
    },
    {
      "id": 203825,
      "baseline": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.4,
        "R": 0.8333,
        "F1": 0.5405
      },
      "precision_prompt": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.32,
        "R": 0.6667,
        "F1": 0.4324
      },
      "added_tags": [
        "asian",
        "black_outfit"
      ],
      "removed_tags": [
        "bat",
        "mammal"
      ]
    },
    {
      "id": 335343,
      "baseline": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.5263,
        "R": 0.6667,
        "F1": 0.5882
      },
      "precision_prompt": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.4783,
        "R": 0.7333,
        "F1": 0.5789
      },
      "added_tags": [
        "palette",
        "resting",
        "serene",
        "sleeping"
      ],
      "removed_tags": []
    },
    {
      "id": 1023509,
      "baseline": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.4286,
        "R": 0.6923,
        "F1": 0.5294
      },
      "precision_prompt": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.4091,
        "R": 0.6923,
        "F1": 0.5143
      },
      "added_tags": [
        "bright_light",
        "feral"
      ],
      "removed_tags": [
        "note"
      ]
    },
    {
      "id": 2141904,
      "baseline": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.2308,
        "R": 0.375,
        "F1": 0.2857
      },
      "precision_prompt": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.2308,
        "R": 0.375,
        "F1": 0.2857
      },
      "added_tags": [],
      "removed_tags": []
    },
    {
      "id": 2291202,
      "baseline": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.5455,
        "R": 0.6667,
        "F1": 0.6
      },
      "precision_prompt": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.6,
        "R": 0.6667,
        "F1": 0.6316
      },
      "added_tags": [
        "fingers",
        "looking_at_viewer"
      ],
      "removed_tags": [
        "gesture",
        "open_mouth",
        "pointing",
        "pointing_up"
      ]
    },
    {
      "id": 3285630,
      "baseline": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.375,
        "R": 0.75,
        "F1": 0.5
      },
      "precision_prompt": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.3929,
        "R": 0.9167,
        "F1": 0.55
      },
      "added_tags": [
        "alpha_channel",
        "brown_clothing",
        "brown_topwear",
        "brown_vest",
        "male"
      ],
      "removed_tags": [
        "business_attire"
      ]
    },
    {
      "id": 260449,
      "baseline": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.5,
        "R": 0.7857,
        "F1": 0.6111
      },
      "precision_prompt": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.5,
        "R": 0.7857,
        "F1": 0.6111
      },
      "added_tags": [
        "raised_arm",
        "topless"
      ],
      "removed_tags": [
        "bottomwear",
        "loincloth"
      ]
    },
    {
      "id": 1325009,
      "baseline": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.65,
        "R": 0.5909,
        "F1": 0.619
      },
      "precision_prompt": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.6522,
        "R": 0.6818,
        "F1": 0.6667
      },
      "added_tags": [
        "bottomwear",
        "gesture",
        "pantherine",
        "pose",
        "raised_hand",
        "shorts",
        "tiger"
      ],
      "removed_tags": [
        "fur_tuft",
        "muscular",
        "thoughtful",
        "tuft"
      ]
    },
    {
      "id": 212271,
      "baseline": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.7222,
        "R": 0.7222,
        "F1": 0.7222
      },
      "precision_prompt": {
        "sel_clothed": true,
        "sel_clothing": true,
        "P": 0.7647,
        "R": 0.7222,
        "F1": 0.7429
      },
      "added_tags": [],
      "removed_tags": [
        "bottomless"
      ]
    }
  ],
  "files": {
    "subset_eval_path": "E:\\image\\backup\\Prompt_Squirrel_RAG\\data\\runtime_debug\\clothing_precision_subset_n10_20260322T113759Z.jsonl",
    "modified_structural_csv": "E:\\image\\backup\\Prompt_Squirrel_RAG\\data\\runtime_debug\\structural_tag_definitions_precision_20260322T113759Z.csv",
    "modified_probe_definitions_csv": "E:\\image\\backup\\Prompt_Squirrel_RAG\\data\\runtime_debug\\probe_tag_definitions_precision_20260322T113759Z.csv"
  }
}